2021/02/10 by Joaquim Silva, Silva, Joaquim, Eduardo R. B. Marques +5
Computer Science · #Age of Information Optimization #Cloud Computing and Resource Management #Distributed #FOS: Computer and information sciences #IoT and Edge/Fog Computing #Parallel #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.2102.05504
openalex publication_date 2021/02/10 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
We present a model for measuring the impact of offloading soft real-time jobs\nover multi-tier cloud infrastructures. The jobs originate in mobile devices and\noffloading strategies may choose to execute them locally, in neighbouring\ndevices, in cloudlets or in infrastructure cloud servers. Within this\nspecification, we put forward several such offloading strategies characterised\nby their differential use of the cloud tiers with the goal of optimizing\nexecution time and/or energy consumption. We implement an instance of the model\nusing Jay, a software framework for adaptive computation offloading in hybrid\nedge clouds. The framework is modular and allows the model and the offloading\nstrategies to be seamlessly implemented while providing the tools to make\ninformed runtime offloading decisions based on system feedback, namely through\na built-in system profiler that gathers runtime information such as workload,\nenergy consumption and available bandwidth for every participating device or\nserver. The results show that offloading strategies sensitive to runtime\nconditions can effectively and dynamically adjust their offloading decisions to\nproduce significant gains in terms of their target optimization functions,\nnamely, execution time, energy consumption and fulfillment of job deadlines.\n